Abstract
The stochastic dynamic inventory-routing problem (SDIRP) is a fundamental problem within supply chain operations that integrates inventory management and vehicle routing while handling the stochastic and dynamic nature of exogenous factors unveiled over time, such as customer demands, inventory supply and travel times. While practical applications require dynamic and stochastic decision-making, research in this field has only recently experienced significant growth, with most inventory-routing literature focusing on static variants. This paper reviews the current state of research on SDIRPs, identifying critical gaps and highlighting emerging trends in problem settings and decision policies. We extend the existing inventory-routing taxonomies by incorporating additional problem characteristics to better align models with real-world contexts. As a result, we highlight the need to account for further sources of uncertainty, multiple-supplier networks, perishability, multiple objectives, and pickup and delivery operations. We further categorize each study based on its policy design, investigating how different problem aspects shape decision policies. To conclude, we emphasize that large-scale and real-time problems require more attention and can benefit from decomposition approaches and learning-based methods.

Abstract
In recent years, the research community, but also the general public, has raised serious questions about the reproducibility and replicability of scientific work. Since many studies include some kind of computational work, these issues are also a technological challenge, not only in computer science, but also in most research domains. Computational replicability and reproducibility are not easy to achieve due to the variety of computational environments that can be used. Indeed, it is challenging to recreate the same environment via the same frameworks, code, programming languages, dependencies, and so on. We propose a framework, known as SciRep, that supports the configuration, execution, and packaging of computational experiments by defining their code, data, programming languages, dependencies, databases, and commands to be executed. After the initial configuration, the experiments can be executed any number of times, always producing exactly the same results. Our approach allows the creation of a reproducibility package for experiments from multiple scientific fields, from medicine to computer science, which can be re-executed on any computer. The produced package acts as a capsule, holding absolutely everything necessary to re-execute the experiment. To evaluate our framework, we compare it with three state-of-the-art tools and use it to reproduce 18 experiments extracted from published scientific articles. With our approach, we were able to execute 16 (89%) of those experiments, while the others reached only 61%, thus showing that our approach is effective. Moreover, all the experiments that were executed produced the results presented in the original publication. Thus, SciRep was able to reproduce 100% of the experiments it could run.

Abstract
The increasing adoption of the Digital Twin concept and technology for managing complex physical assets has led to the emergence of Digital Twin Ecosystems, where interconnected digital twins generate additional value. However, ensuring seamless data sharing and interoperability among diverse systems presents significant challenges. Although research on digital twin architectures has advanced, gaps remain in addressing data governance, security, and stakeholders' trust. This study performs a comprehensive literature review to investigate architectural solutions to overcome challenges in digital twin ecosystems. The findings identify key requirements such as interoperability, governance, and data management, emphasizing the role of Data Spaces as enablers of secure data sharing. By structuring the requirements for digital twin ecosystem architectures, this paper identifies gaps suggesting future research on scalable and sustainable digital twin ecosystem implementations. These insights are expected to contribute to the development of frameworks that integrate technical advances with organizational and regulatory considerations, ultimately fostering the adoption of digital twin ecosystems across industries.

Abstract
[No abstract available]

Abstract

Abstract
Machine learning (ML) frameworks, such as PyTorch and TensorFlow, rely on data loaders to preprocess data before feeding it to accelerators. When preprocessing is inefficiently pipelined, GPUs can remain idle over long periods of time, leading to substantial training delays. For example, PyTorch’s default data loaders can cause up to 76% GPU idleness. A key bottleneck is the variability in preprocessing time across samples within the same dataset. Existing data loaders are oblivious to this variability, training all samples uniformly. In this case, a single slow sample can stall the entire batch, causing head-of-line blocking. We present MinatoLoader, a general-purpose data loader for PyTorch that accelerates training and improves GPU utilization under single-server, multi-GPU settings. It continuously prepares data in background and constructs batches by prioritizing fast-to-process samples, while slower samples are processed in parallel. Experiments conducted over NVIDIA V100 and A100 GPUs show that MinatoLoader accelerates training by up to 7.5× (3.6× on average) over PyTorch DataLoader and Pecan, and up to 3× (2.2× on average) over DALI. It also increases average GPU utilization from 46% with PyTorch to 90%, while preserving model accuracy and enabling faster convergence. © 2026 Copyright held by the owner/author(s)